Kim Sewon, Jang Hanbyol, Jang Jinseong, Lee Young Han, Hwang Dosik
School of Electrical and Electronic Engineering, Yonsei University, Seoul, Korea.
Department of Radiology and Center for Clinical Imaging Data Science (CCIDS), Yonsei University College of Medicine, Seoul, Korea.
Magn Reson Med. 2020 Dec;84(6):2994-3008. doi: 10.1002/mrm.28327. Epub 2020 Jun 1.
To generate short tau, or short inversion time (TI), inversion recovery (STIR) images from three multi-contrast MR images, without additional scanning, using a deep neural network.
For simulation studies, we used multi-contrast simulation images. For in-vivo studies, we acquired knee MR images including 288 slices of T -weighted (T -w), T -weighted (T -w), gradient-recalled echo (GRE), and STIR images taken from 12 healthy volunteers. Our MR image synthesis method generates a new contrast MR image from multi-contrast MR images. We used a deep neural network to identify the complex relationships between MR images that show various contrasts for the same tissues. Our contrast-conversion deep neural network (CC-DNN) is an end-to-end architecture that trains the model to create one image from three (T -w, T -w, and GRE images). We propose a new loss function to take into account intensity differences, misregistration, and local intensity variations. The CC-DNN-generated STIR images were evaluated with four quantitative evaluation metrics, including mean squared error, peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and multi-scale SSIM (MS-SSIM). Furthermore, a subjective evaluation was performed by musculoskeletal radiologists.
Our method showed improved results in all quantitative evaluations compared with other methods and received the highest scores in subjective evaluations by musculoskeletal radiologists.
This study suggests the feasibility of our method for generating STIR sequence images without additional scanning that offered a potential alternative to the STIR pulse sequence when additional scanning is limited or STIR artifacts are severe.
使用深度神经网络从三张多对比度磁共振（MR）图像生成短反转时间（TI）反转恢复（STIR）图像，无需额外扫描。
对于模拟研究，我们使用了多对比度模拟图像。对于体内研究，我们采集了12名健康志愿者的膝关节MR图像，包括288层T加权（T-w）、T加权（T-w）、梯度回波（GRE）和STIR图像。我们的MR图像合成方法从多对比度MR图像生成新的对比度MR图像。我们使用深度神经网络来识别同一组织在不同对比度下的MR图像之间的复杂关系。我们的对比度转换深度神经网络（CC-DNN）是一种端到端架构，训练该模型从三张图像（T-w、T-w和GRE图像）创建一张图像。我们提出了一种新的损失函数，以考虑强度差异、配准误差和局部强度变化。使用包括均方误差、峰值信噪比（PSNR）、结构相似性（SSIM）和多尺度SSIM（MS-SSIM）在内的四个定量评估指标对CC-DNN生成的STIR图像进行评估。此外，由肌肉骨骼放射科医生进行主观评估。
与其他方法相比，我们的方法在所有定量评估中均显示出更好的结果，并且在肌肉骨骼放射科医生的主观评估中获得了最高分。
本研究表明我们的方法在不进行额外扫描的情况下生成STIR序列图像的可行性，当额外扫描受限或STIR伪影严重时，该方法为STIR脉冲序列提供了一种潜在的替代方案。
